Starch/fiber/poly(lactic acid) foam and compressed foam composites

Abstract

Composites of starch, fiber, and poly(lactic acid) (PLA) were made using a foam substrate formed by dehydrating starch or starch/fiber gels. PLA was infiltrated into the dry foam to provide better moisture resistance. Foam composites were also compressed into plastics using force ranging from 4–76 MPa. Tensile strength increased with increasing compression force applied to the foam sample. The samples became increasingly transparent with compression forces approaching 76 MPa. PLA infusion into starch and starch/fiber foam composites resulted in PLA content of 20% and 33%, respectively and provided moisture resistance to the outer regions of the foam samples. The PLA-infused foam samples increased in tensile strength when compressed up to 29 MPa. The PLA-infused compressed samples had greater moisture resistance and had intermediate rates of mineralization compared to the control samples.

---

Introduction

Starch is a semi-crystalline polymer composed of amylose and amylopectin molecules.¹ Amylose is primarily a linear polymer composed of about 1000 (1–4)-linked α-D-glucopyranosyl units.² Amylopectin is a very large molecule (105–106 glucose units) made up of (1–4)-linked α-D-glucopyranose chains with α-(1–6) branches.^1,2 Starch is one of the most abundant natural polymers and has many of the properties sought for in single-use, disposable packaging.^2,3 It is cheap, readily degradable in composting conditions, and, properly plasticized, starch can be processed as a thermoplastic similar to commodity plastics and made into commercial products.^4,5 Despite broad interest in starch-based plastics, they have only found use in various niche markets. This is due primarily to their sensitivity to moisture and poor mechanical properties.

One way to improve the mechanical properties of starch plastics is to incorporate reinforcing fiber such as pulped plant fiber.⁶ Pulped fibers may be readily dispersed in a starch matrix to improve tensile strength and flexibility of starch plastics, extruded foam, and films.^6,7 Starch/fiber mixtures have also been made into a lightweight foam composite using a baking process.^3,8 The baked article may be laminated with a moisture resistant polyester film that, in concert with the fiber, helps improve the strength, flexibility, and moisture resistance of the composite.^3,8,9

Starch composites with improved moisture resistance have also been reported for blends of starch and other materials such as acrylic acid,¹⁰ poly(vinyl alcohol),¹¹ ethylene-vinyl acetate copolymer,¹² poly(caprolactone)¹³ and poly(lactic acid) (PLA) among others. PLA is of particular interest in composite materials because it is a biodegradable thermoplastic derived from renewable sources, such as starch and sugar. It is an aliphatic polyester made up of lactic acid (2-hydroxy propionic acid).¹⁴ PLA is synthesized by the condensation polymerization of L- or D-lactic acid or ring-opening polymerization of lactides.^15,16 Successful industrial-scale production has resulted in PLA becoming more cost competitive with polyolefins. It is the first biobased polyester to be classified as a commodity plastic and a viable alternative to petroleum-based plastics for packaging materials.^17,18 The use of low cost fillers and additives in making low cost, functional PLA composites continues to be an active area of research.

Composite materials made of starch and PLA have been investigated for many years.^19–26 Many starch/PLA composites are made using an extrusion process wherein the PLA forms the continuous phase of the matrix while the starch component serves as a filler.^19,25,26 PLA and starch are thermodynamically immiscible and interfacial fracture strength is low due to the low interdiffusion of starch and PLA molecules.^19,20 Interfacial fracture strength may be increased by using chemical compatibilizers but these may significantly increase the cost and raise concerns associated with chemical use and exposure.¹⁹ The use of starch as a filler generally increases the rigidity and brittleness of PLA compared to the neat polymer and reduces tensile strength.^24,26

Fiber is also used as a low-cost, renewable material for making PLA composites. Many PLA/fiber composites are made by a compression process, especially when longer fibers or very high fiber content are used.^23,27 Commercial products including cell phone shells and automobile parts are made using compression molded PLA/fiber composites.²⁷ Starch/fiber/PLA blends that are compatible with the compression molding process could provide a way of making high fiber composites with improved mechanical properties and higher rates of biodegradation.

Although very few starch/fiber composites have been reported using compression molding, it has been used in making starch plastics from starch foam.²⁸ The foam is made by dehydrating aqueous starch gels using ethanol and then drying into a porous matrix with extremely small (<2 μm), open pores.²⁹ Starch/fiber foam composites have also been reported using the solvent dehydration process.³⁰ The composite foam may be compression molded into fiber reinforced starch composites with improved mechanical properties.³⁰ However, moisture sensitivity remains a limitation of such starch and starch/fiber plastics. PLA could provide a moisture barrier for such plastics if applied as a laminate film, for instance. Another approach is to infiltrate the porous matrix of starch or starch/fiber foam with a solution of PLA solubilized in a solvent. Evaporation of the solvent would leave a residue of PLA film coating the exposed surfaces within the porous matrix of starch/fiber composites and provide moisture resistance. This process could be useful for making moisture resistant starch/PLA and starch/fiber/PLA foam composites. It is not known how the infusion of PLA might affect the processability of composite foams into plastics by compression molding or whether such plastic composites will have useful mechanical properties and moisture resistance. The purpose of this study was to investigate the properties of starch/PLA and starch/fiber/PLA foam materials and plastics made by a compression molding process.

Materials and methods

Unmodified wheat starch (Midsol 50, Midwest Grain Products, Atchison, KS) was used to produce samples of starch foam and subsequently their plastics. The wheat starch was composed of about 28% amylose and 72% amylopectin. Poly(DL-lactide) (mol. wt = 80–100 kDa) was supplied by NatureWorks® (Ingeo™ Biopolymer 3052D). Pulped Southern bleached softwood fiber (SWF, Leaf River 90, moisture content 7.2%, fiber length 2.47 mm) was provided by Georgia Pacific, Atlanta, GA.

Starch foam preparation

Starch samples were prepared from an aqueous starch slurry (8% w/w) that was heated in a boiling water bath. In samples containing fiber (5% w/w), the fiber fraction was first combined with the water and allowed to hydrate overnight before being dispersed by rigorously mixing. The starch fraction was added to the mixture and the contents were stirred constantly while heating in a boiling water bath. Viscosity was monitored (Brookfield, model RVT, Stoughton, MA) to determine when peak viscosity of the gelatinized starch melt was reached. The starch melt was poured into molds (35.0 × 43.0 × 1.9 cm) and refrigerated overnight to allow gelation to occur. The starch gels were removed from the molds and soaked 24 h in successive baths of ethanol (40, 70, 90% and three changes of 100% ethanol). The fully dehydrated starch slabs were approximately 60% of their original volume due to shrinkage whereas the starch/fiber slabs had less shrinkage (95% of their original volume). The dehydrated gels were removed from the last ethanol bath and placed on perforated trays covered with sheets of filter paper. The trays were placed in a forced-air drier (60 °C) overnight to remove the ethanol. The dry foam samples comprised of either starch (100%) or starch/fiber (62% and 38%, respectively) were stored in plastic bags until used.

Microscopy

Light micrographs of starch/PLA foam composites were performed using a camera-mounted stereomicroscope (Leica, Model MZ 16F, Wetzlar, Germany). A double edged razor blade was used to make hand sections approximately 1–2 mm in thickness. The sections were mounted on a glass slide. Several drops of water were deposited directly on the sections before a cover slip was mounted on the sample.

Scanning electron micrographs (SEM)

Samples were mounted with adhesive tape on aluminum stubs. Cross-sections of fractured samples were mounted with the cross-section positioned upward on aluminum stubs. Failure analysis of tensile bars was performed by imaging the fracture surface of failed tensile bars. All specimens were sputter-coated with gold–palladium for 45 s in a Denton Desk II sputter coating unit (Denton Vacuum USA, Moorestown, NJ). Specimens were analyzed using a Hitachi S4700 field emission scanning electron microscope (Hitachi HTA, Japan) at 2 kV.

PLA infiltration

A solution of PLA (7.5% w/v) in chloroform was prepared by solubilizing PLA pellets in chloroform in a sealed container at 40 °C. The PLA solution was cooled to room temperature and poured in a dish. Starch or starch/fiber foam samples were placed in the dish and submerged completely in the PLA solution by using weights to keep them from floating. The dish was then placed in a vacuum chamber and subjected to vacuum (100 Torr) for 1 min to evacuate air within the samples. PLA solution infiltrated the foam samples upon release of the vacuum. This process was repeated a second time to insure thorough infiltration of the PLA solution. The samples were removed from the vacuum chamber, blotted and dried in a forced air oven (60 °C) for 4 h. Starch/PLA and starch/fiber/PLA plastic composites were made by compressing PLA-infiltrated foam (at 29 MPa) samples between two steel plates heated to 160 °C. The plastic samples were then conditioned and mechanically tested as per ASTM D638-10.

Water immersion tests

Water immersion tests were performed as per ASTM D570-98. Test specimens (25.4 × 1.27 × 0.2 cm) were dried at 110 °C in an oven for 24 h then immediately stored in a desiccator prior to testing. The samples were removed from the desiccator and immersed in water at 25 °C ± 2 °C for a period of 24 h. Periodically during the immersion test, the samples were blotted and weighed to monitor the rate of moisture absorption. The mean values from three replicates were reported for foam and plastic composite specimens.

X-ray diffraction (XRD)

The X-ray diffraction patterns for neat starch and compressed samples were obtained with an X-ray diffractometer (VEB Carl Zeiss-Jena URD-6), using Cu Kα radiation (λ = 1.5406 A) at 40 kV and 20 mA. Scattered radiation was detected in the range of 2θ from 5° to 40°, at a scan rate of 2° min⁻¹.

Respirometry

The relative degradation rate of samples was determined using a respirometer (Micro-Oxymax System, Columbus Instruments, Columbus, OH). The respirometer CO₂ sensor was calibrated with a CO₂ standard gas (8000 ppm). Carbon content of the samples was determined using a CHN elemental analyzer (Perkin Elmer 2400, Boston, MA). The analyzer was equipped with a thermoconductivity detector and was operated using helium gas. The combustion temperature was 975 °C and the reduction temperature was 680 °C. Commercial compost was purchased locally and adjusted to 55% moisture (dry wt basis). The glass sample bottles (250 ml) were filled with 20.0 g of compost and 0.30 g of test sample that was milled and gently mixed with the compost. The sample bottles were initially flushed with CO₂-free air and sealed. Respirometry experiments were conducted at room temperature (22 °C) and CO₂ concentration was read every 12 h. Three replications were tested for each treatment and data were expressed in terms of percent mineralization.

Mechanical tests

Starch and starch/fiber foam samples. Dry foam slabs were removed from plastic bags and a template of a type I tensile bar (165 mm × 19 mm as per ASTM D638-10) was traced on the slab surface. Tensile bars were cut using a band saw and then conditioned at 25 ± 1 °C and 50% RH for at least 48 h before performing tensile tests. Tensile bars were deformed in tension at a rate of 5 mm min⁻¹ using a universal testing machine (Instron 4500, Canton, MA). To make plastic samples from starch and starch/fiber foam, tensile bars (also type I) were cut on a band saw, conditioned and compressed between two polished plates using forces ranging from 4.0–76 MPa. Tensile properties were determined as per ASTM D638-10 using a minimum of five replicates. Compression tests were performed only on the foam samples as per ASTM D1621-10 except that compression strength was determined at 20% deformation. Tests were performed using a deformation rate of 5 mm min⁻¹.

Results and discussion

Sample characterization

The starch used in the present study consisted of starch granules with a bimodal size distribution as previously reported by Svihus et al.³¹ The larger granules were lenticular in shape (Fig. 1A) with diameters in the range of 16–26 μm. The smaller granules were oval in shape with diameters ranging from 2–10 μm. The continuous starch matrix formed by the gelatinization and dehydration process was white, opaque, and low in density (Table 1). Starch/fiber foam had lower density than starch foam. The difference in density can be attributed to the greater shrinkage and densification of the starch foam samples (ca. 60%) versus the starch/fiber foam (ca. 5%) during the dehydration process. It is well known that higher foam density generally leads to greater tensile strength and modulus.³² However, the starch/fiber foam had higher tensile strength, modulus, and elongation to break than the starch foam in spite of its lower density underscoring the effect of the fiber in improving tensile properties.

Fig. 1 (A) SEM view of granular wheat starch; (B) starch foam matrix with no compression treatment. Arrows indicate starch granule remnants; (C, D, and E) cross sectional view of fractured surface of foam samples after compression with 7, 29 and 76 MPa force, respectively. Arrow indicates presence of starch granule remnants within the fractured matrix of the sample; (F) photograph illustrating the transparency of sample compressed to 76 MPa; (G and H) failure zone of tensile bars following tensile tests for starch fiber composites compressed to 7 and 76 MPa, respectively.

Table 1 Density, compressive and tensile properties of uncompressed starch and starch/fiber foams with and without poly(lactic acid) infiltration^a

Sample	Density (g cm^−3)	Tensile strength (MPa)	Elastic modulus (MPa)	Elongation at break (%)	Compressive strength (MPa)
a * Values followed by a different letter are significantly different (0.05 level). ** Average PLA content was 20%. *** Average PLA content was 33%.
Starch foam	0.318a*	0.24a	17.8a	2.38a	1.18a
Starch/fiber	0.247b	0.41b	25.3b	4.75b	0.46b
Starch/PLA**	0.380c	0.27a	39.7c	1.36c	2.11c
Starch/fiber/PLA***	0.362c	1.00c	47.6c	5.41b	2.54d

---

The microstructure of the starch matrix consisted of a porous, fibrous network of starch with remnants of starch granules interspersed throughout (Fig. 1B). Other processes such as freeze-drying and extrusion may also be used to form a porous starch matrix. However, these processes do not create the very fine micro-domains produced by the solvent dehydration process.²⁹

Cross-sectional views of starch samples compressed at 7 and 29 MPa revealed that the matrix was more compact and dense compared to the control but still opaque and somewhat porous (Fig. 1C and D). The opaque nature of the compressed and control foam samples was likely due to the small pores interspersed throughout the matrix. Pore sizes greater than the wavelength of visible light tend to scatter light and lead to opacity. Pores smaller than 0.4 μm are needed for transparency.³³ Foam samples compressed at 76 MPa were dense with very little porosity and became transparent (Fig. 1E and F). Some small pores could still be seen in the matrix of foams compressed at 76 MPa (Fig. 1E). However, the pores were sufficiently small to render the sample transparent (Fig. 1F).

PLA infiltration

The primary purpose of the PLA infiltration treatment was to confer greater moisture resistance to the starch and starch/fiber composites. Starch-based materials are highly hydrophilic which makes them moisture sensitive and susceptible to mechanical and structural changes when exposed to water (Fig. 2A). Such changes make starch-based materials unstable and unsuitable for many applications. Hand sections (thickness ≅ 1 mm) of starch foam quickly saturated with water and became a clear gel with very little mechanical strength (Fig. 2A). The distribution of PLA within the matrix of starch/PLA foam samples was revealed in hand sections exposed to water (Fig. 2B). The core region of the dry starch/PLA foam sample contained very little PLA and was susceptible to water absorption and swelling (Fig. 5A). The outer regions of the PLA infiltrated starch foam was protected against moisture exposure due to the well dispersed distribution of PLA. As previously mentioned, the starch/fiber foam was more porous and was infused with higher amounts of PLA. These samples had PLA distributed throughout the entire matrix, including the core region (data not shown).

Fig. 2 Photo of hand sections of starch foam (A) and (B) PLA-infiltrated starch foam exposed to water. The core region of the PLA-infiltrated sample contained very little PLA and swelled when placed in contact with water. (C) SEM micrograph of interior region of starch foam; (D) SEM micrograph of interior region of PLA-infiltrated starch foam. Inset-SEM micrograph near to surface of PLA-infiltrated starch foam.

SEM Micrographs of the dry foam with and without PLA were taken to reveal the distribution of PLA within the foam matrix (Fig. 2C and D). The starch matrix was composed of a network of fine strands or fibers with some starch granule remnants (compare Fig. 2C and 1B). The PLA appeared to be deposited as a fine coating on the surface of the fine starch fibers (Fig. 2D). Fibers that were near the foam surface were comprised of relatively course fibers due to the thick coating of PLA (Fig. 2D, see inset).

Infiltrating the starch and starch/fiber foams with PLA resulted in PLA concentrations of 20% and 33%, respectively (Table 1). The higher PLA content in starch/fiber foams was due to the greater porosity that allowed greater infiltration of the PLA solution and higher weight gain during the infiltration process compared to the starch foam. As a result of the greater PLA infiltration, the starch/fiber/PLA foam increased in density more than the PLA infiltrated starch foam (46% versus 20%, respectively) and had the highest tensile and compressive strength. The PLA infiltration made the starch foam more rigid and brittle (Table 1).

Tensile failure analysis

The effect of fiber on mechanical properties of the starch samples was reflected in force–deformation curves obtained from tensile tests (Fig. 3). Foam samples compressed with 76 MPa force required greater force to achieve tensile failure compared to samples compressed with 7 MPa force (Fig. 3). Both of the starch samples were brittle and required little deformation to induce tensile failure. The starch/fiber samples required greater deformation than the starch samples to incur tensile failure, especially for the samples compressed with 7 MPa force (Fig. 3). The greater elongation to break in starch/fiber composites may be due to both the reinforcing effect of fibers in the starch matrix⁶ and the effect of fiber entanglement. Fiber entanglement is dependent on fiber concentration and aspect ratio. The aspect ratio of the softwood fiber used in the present study was high (∼100). The high fiber content of fibers with a high aspect ratio were conducive to fiber entanglement. Fiber entanglement allows the sample to gradually fail as the fibers pull apart. Failure analyses of tensile bars from starch/fiber samples compressed at 7 and 76 MPa were performed by SEM to better understand the failure mechanism (Fig. 1G and H). The sample compressed at 7 MPa (Fig. 1G) had a dense distribution of fibers in the failure zone and extensive fiber pull-out. It appeared that along with the fiber pull-out the starch matrix may have also disintegrated around the failure zone. Very little of the matrix material remained adhered to the loose fibers. It is likely that the starch matrix material failed around the fibers as they stretched thus freeing the fibers.

Fig. 3 Force–deformation curves for foam samples compressed at 7 or 76 MPa of force, tested in tension.

The failure zone of samples compressed at 76 MPa was a stark contrast to that of samples compressed at 7 MPa (Fig. 1H). There was little or no fiber pull-out in the failure zone (Fig. 1H). Instead of the fibers pulling out, they broke cleanly along the same plane of the fractured starch matrix. The results indicate better interfacial adhesion between the starch and fiber components in samples compressed at 76 MPa. The greater compressive force had the effect of removing void spaces in the matrix thus increasing the contact area between fibers and the starch matrix. The increase in contact area likely increased the amount of hydrogen bonding between the starch and fiber which could account for the greater tensile strength observed.

X-Ray diffraction

Native wheat starch, compressed starch foam (76 MPa), and compressed starch/fiber foam (76 MPa) gave distinct diffractograms in X-ray diffraction studies (Fig. 4). The native wheat starch had a semi-crystalline structure with peaks typical of A-type crystals (2θ ∼ 14.8°, 17.0–18.0°, 19.9° and 23.5°).³⁴ The crystallinity was lost during the gelatinization/compression processes resulting in diffractograms exhibiting a broad amorphous peak with no significant residual A-type crystallinity. The results indicate that the starch granule structure was disrupted by the processes of gelatinization and high compression (Fig. 1E). Small peaks were found around 2θ ∼ 12.0°, 17.3° and 19.5°, which is characteristic for crystallographic parameters of V_H-type crystals for destructured starch prepared from solution.³⁴ For the starch/fiber sample, a typical cellulose type 1 profile³⁵ was obtained indicating that the crystallinity of the fiber overlaps with the gelatinized starch.

Fig. 4 X-ray diffraction profile for native starch, starch (76 MPa) and starch/fiber (76 MPa).

X-Ray diffractograms of PLA and PLA composites were studied to better understand their crystalline properties. The diffractograms showed peaks typical of the PLA profile (2θ = 16.8° and 19.5°) were present in both the starch/PLA and starch/fiber/PLA samples (Fig. 5). A peak found in the fiber composite (2θ = 22.6°) was consistent with cellulose type I fiber. Additionally, gelatinized starch can be seen through the amorphous halo in the 2θ region of 10° to 30° (compare Fig. 4 and 5).

Fig. 5 X-ray diffraction profile for neat PLA, starch/PLA (29 MPa) and starch/fiber/PLA (29 MPa).

Mineralization

One of the unique properties of starch-based materials is their rapid rate of biodegradation. Baked starch-based single-use food service packaging degrades in a composting environment in about five weeks. The rapid rate of starch biodegradation makes it attractive for single-use food service packaging. While PLA is also biodegradable, its rate of mineralization at room temperature is much lower than starch (Fig. 6). By combining starch with PLA, composite materials with improved moisture resistance have been made that still maintain high rates of mineralization (Fig. 6). The mineralization rate for the starch/PLA sample was intermediate as expected since the starch composite was comprised of 20% PLA. The compressed starch/fiber/PLA had a lower mineralization rate than the compressed starch/PLA probably because of its higher PLA content (33%) and the cellulose fiber component.

Fig. 6 Mineralization rates of starch, starch/fiber composites, starch/PLA composites, starch/fiber/PLA composites, and neat PLA.

Mechanical properties

Foam samples. Although the mechanical properties of the samples were tested under tensile deformation, there should be no mistaking that the samples themselves were prepared using different amounts of compression force. The effect of compression force on the tensile strength of foam samples was determined (Fig. 7). The foam samples containing no PLA (starch, starch/fiber) were compressed at room temperature and over a broader range of force than the PLA foam samples. Tensile strength increased at a lower rate than with the PLA samples but generally increased up to 76 MPa of compression force (Fig. 7). Initially, the tensile strength increased linearly as compression force increased. Tensile strength of samples prepared using compression forces greater than 29 MPa continued to increase but at a lower rate. Starch/fiber plastic composites made from compressed foam samples generally had higher tensile strength than samples without fiber (Fig. 7). However, the difference in tensile strength was not significant at the 29 MPa force.

Fig. 7 Tensile strength of starch, starch/fiber, starch/PLA and starch/fiber/PLA prepared using different compression force. The PLA infused samples were compressed with heated (160 °C) platens.

PLA samples. The PLA foam samples (starch/PLA and starch/fiber PLA) which were compressed with heated platens (160 °C) increased in tensile strength as compression force used in preparing the samples was increased up to approximately 14 MPa (Fig. 7). No further increase in tensile strength was attained by using greater compression force. To the contrary, the tensile strength of starch/PLA composites decreased when greater compression force was applied. The decrease in tensile strength may be due to the PLA component interfering with the compaction and interaction of the continuous starch matrix during the compression treatment. The tensile strength of the starch/fiber/PLA composite prepared using 14 MPa compression force was nearly double the tensile strength of the PLA composite without fiber. The starch/PLA and starch/fiber/PLA composites compressed using 14 MPa were transluscent similar to starch foam samples compressed to 76 MPa (data not shown).

A compression force of 29 MPa (marked in Fig. 7) was arbitrarily selected for making comparisons of the tensile properties of the compressed foam samples (Table 2). The effect of PLA on the properties of compressed foam samples was variable. Starch/PLA composites with no fiber had very low tensile strength and were brittle as evident by the low elongation to break (Table 2). The starch/fiber/PLA composite had the highest tensile strength, elongation to break, and toughness of all the samples tested at the 29 MPa compression treatment. Interestingly, the samples containing PLA required less compressive force to reach maximum tensile strength (Fig. 7).

Table 2 Tensile properties of compressed (29 MPa) foam samples^a

Sample	Tensile strength (MPa)	Elongation to break (%)	Elastic modulus (GPa)	Toughness (MPa)
a * Values within columns followed by a different letter are significantly different (0.05 level).
Starch	17.7a*	2.27a	1.06a	0.22a
Starch/fiber (95 : 5)	18.9a	2.37a	1.34b	0.39a
Starch/PLA (80 : 20)	3.16b	1.48a	0.282c	0.036a
Starch/fiber/PLA (62 : 5 : 33)	25.3c	7.23b	0.623d	1.15b

---

There was no benefit in compressing the starch/fiber/PLA samples more than 29 MPa or the starch/PLA sample more than 14 MPa (Fig. 7). The reason lower compression force was needed to maximize the strength of the PLA composites may be due to the addition of heat during the compression treatment used to form the plastic. Heated platens soften or even melt the PLA, which in turn, may have facilitated the compression of the foam structure.

Ke and Sun²⁶ showed that adding granular starch as a filler in a continuous PLA matrix markedly reduced tensile strength and elongation to break while increasing modulus (Table 3). The starch/PLA composite in the present study contained about 80% starch and when compressed at 14 MPa (Fig. 7) was comparable in strength to that of a comparable starch/PLA composite reported by Ke and Sun (Table 3). The starch/fiber/PLA composite in the present study contained about 30% PLA and had higher tensile strength compared to a 70 : 30 blend of starch : PLA previously reported (compare Tables 2 and 3). The results support the role of fiber in improving the properties of starch/PLA composites.

Table 3 Tensile properties of starch/PLA composites using starch granules as filler and PLA as continuous matrix. Data from Ke and Sun²⁶

Sample	Tensile strength (MPa)	Elongation to break (%)	Elastic modulus (GPa)
PLA	61	6.33	1.204
Starch/PLA (80 : 20)	13.2	0.84	1.739
Starch/PLA (70 : 30)	18.9	1.1	1.832

---

Moisture resistance

The results of the water absorption tests for the foam (Fig. 8A) and compressed foam (Fig. 8B) samples show the marked effect PLA has on moisture absorption. In the foam samples, the effect of PLA was most pronounced on the starch/fiber sample (Fig. 8A). The rate of moisture absorption was approximately six fold higher in the starch/fiber foam compared to the starch/fiber/PLA foam. The effect of PLA infusion was less pronounced in the starch foam without fiber. This was likely due in part to the lower amount of PLA infused in the starch foam versus the starch/fiber foam (20% and 33%, respectively). It is also likely due to the lower porosity of the starch foam and the tendency of starch to quickly swell when in contact with water. Such swelling may help seal the pores and slow the absorption of water.

Fig. 8 Weight increase during water immersion test for (A) foam samples and (B) compressed samples.

The starch and starch/PLA compressed foam samples generally had a lower rate of water absorption compared to the foam samples except for the starch/PLA sample which was comparable for both the foam and plastic samples (compare Fig. 8A and B). The compressed starch/fiber/PLA sample had less than half the weight gain of the compressed starch/fiber sample.

The results of this study show that the PLA infiltration of the starch foam matrix decreased the strength of compression molded plastics. This was most likely due to the interference of infused PLA on the hydrogen bonding of starch within the matrix. In contrast to the starch/PLA plastics, the starch/fiber/PLA plastics had strength similar to the starch/fiber plastics. In both of the fiber composites, the interfacial adhesive strength was sufficient to transfer tensile stress to the fibers. This was confirmed by the fact there was no fiber pullout observed in either sample. The results also show that infiltration of starch and/or starch/fiber foams with polymer solutions is effective in providing moisture resistance both for foam and compression molded composites. The infiltration process disperses the PLA throughout the matrix which negates any concerns about delamination that can be an issue with film laminates. There may be other effective ways to use PLA to provide moisture resistance for a porous starch or starch/fiber matrix such as by incorporating a PLA powder or PLA fibers in the matrix during preparation. The composite could then be heat treated to melt the PLA and allow it to infiltrate the porous matrix without the use of a solvent. However, PLA is a hard material and is difficult to cryomill into a powder fine enough to be used in such an approach. Furthermore, the viscosity and surface tension of a PLA melt may preclude it from adequately infiltrating the matrix. The same issues would be encountered in trying to melt PLA fibers dispersed in a starch matrix.

Conclusions

PLA infusion of starch and starch/fiber foams is effective in making moisture resistant composites with higher biodegradation rates than neat PLA. Compression molding of starch/PLA and starch/fiber/PLA foams can be used successfully to make moisture resistant plastic composites with useful mechanical properties. Compression molding, which is well suited to processing composites with high fiber content or greater fiber length could be used for making a wide array of composites with properties suitable for many applications.

References

1. D. French, in Starch: Chemistry and Technology, ed. R. L. Whistler, J. N. Bemiller and E. F. Paschall, Academic Press Inc., Orlando, 2nd edn, 1984, pp. 183–247.
2. R. L. Whistler and J. R. Daniel, in Starch: Chemistry and Technology, ed. R. L. Whistler, J. N. Bemiller and E. F. Paschall, Academic Press, New York, 2nd edn, 1984, pp. 1–9.
3. G. M. Glenn, W. J. Orts and G. A. R. Nobes, Ind. Crops Prod., 2001, 14, 201–212.
4. P. A. Altieri and N. L. Lacourse, US Patent #4,863,655, 1989.
5. R. L. Shogren, G. F. Fanta and W. M. Doane, Starch/Staerke, 1993, 45, 276–280.
6. A. A. S. Curvelo, A. J. F. de Carvalho and J. A. M. Agnelli, Carbohydr. Polym., 2001, 45, 183–188.
7. M. Wollerdorfer and H. Bader, Ind. Crops Prod., 1998, 8, 105–112.
8. P. J. Andersen and S. K. Hodson, US Patent #5,679,145, 1997.
9. G. M. Glenn, W. J. Orts, G. A. R. Nobes and G. M. Gray, Ind. Crops Prod., 2001, 14, 125–134.
10. R. L. Shogren, A. R. Thompson, F. C. Felker, R. E. Harry-Ókuru, S. H. Gordon, R. V. Greene and J. M. Gould, J. Appl. Polym. Sci., 1992, 44, 1971–1978.
11. N. Follain, C. Joly, P. Dole and C. Bliard, Carbohydr. Polym., 2005, 60, 185–192.
12. V. A. Alvarez and A. Vazquez, Polym. Degrad. Stab., 2004, 84, 13–21.
13. L. Averous, L. Moro, P. Dole and C. Fringant, Polymer, 2000, 41, 4157–4167.
14. M. Murariu, A. l. Da Silva Ferreira, P. Degée, M. Alexandre and P. Dubois, Polymer, 2007, 48, 2613–2618.
15. R. Mehta, V. Kumar, H. Bhunia and S. N. Upadhyay, J. Macromol. Sci., Polym. Rev., 2005, 45, 325–349.
16. D. Garlotta, J. Polym. Environ., 2001, 9, 63–84.
17. L. T. Lim, R. Auras and M. Rubino, Prog. Polym. Sci., 2008, 33, 820–852.
18. S. Mecking, Angew. Chem., Int. Ed., 2004, 43, 1078–1085.
19. J.-F. Zhang and X. Sun, Biomacromolecules, 2004, 5, 1446–1451.
20. H. Wang, X. Sun and P. Seib, J. Appl. Polym. Sci., 2001, 82, 1761–1767.
21. R. Shogren, G. Selling and J. L. Willett, J. Polym. Environ., 2011, 19, 329–334.
22. T. Ke and X. Sun, J. Appl. Polym. Sci., 2003, 89, 1203–1210.
23. M. S. Huda, L. T. Drzal, A. K. Mohanty and M. Misra, Compos. Sci. Technol., 2008, 68, 424–432.
24. M. A. Huneault and H. Li, Polymer, 2007, 48, 270–280.
25. S. Jacobsen and H. G. Fritz, Polym. Eng. Sci., 1996, 36, 2799–2804.
26. T. Ke and X. Sun, Cereal Chem., 2000, 77, 761–768.
27. N. Graupner, A. S. Herrmann and J. Mussig, Composites, Part A, 2009, 40, 810–821.
28. G. M. Glenn and J. Hsu, Ind. Crops Prod., 1997, 7, 37–44.
29. G. M. Glenn and D. W. Irving, Cereal Chem., 1995, 72, 155–161.
30. G. M. Glenn, W. J. Orts, R. G. Buttery and D. Stern, in Biopolymers from Polysaccharides and Agroproteins, ed. R. A. Gross and G. F. Leatham, ACS, Washington, DC., 2001, pp. 102–116.
31. B. Svihus, A. K. Uhlen and O. M. Harstad, Anim. Feed Sci. Technol., 2005, 122, 303–320.
32. M. E. Kabir, M. C. Saha and S. Jeelani, Mater. Sci. Eng., A, 2006, 429, 225–235.
33. H. Yabu and M. Shimomura, Chem. Mater., 2005, 17, 5231–5234.
34. J. J. G. van Soest, S. H. D. Hulleman, D. de Wit and J. F. G. Vliegenthart, Ind. Crops Prod., 1996, 5, 11–22.
35. D. Klemm, B. Heublein, H.-P. Fink and A. Bohn, Angew. Chem., Int. Ed., 2005, 44, 3358–3393.

---